Deep neural networks (DNNs)-based battery remaining useful life (RUL) prognostics models suffer from high computational costs, long inference times, and a tendency to overfit, posing significant challenges for deployment on memory-constrained edge devices. To address this, LDNet-RUL, a lightweight yet high-performance model, is proposed for lithiumion battery lifetime prognostics. Specifically, a deformable depthwise convolution unit is designed that efficiently adapts to variations in input features with relatively low computational cost, which leverages fewer parameters and an adaptive receptive field to capture battery degradation dynamics. A multi-scale feature extraction method is introduced to extract degradation features from both short-term and long-term data during the battery charging process, which makes the method applicable to various battery charging protocols. To address overfitting and accelerate model inference, an adaptive channel selector, an adaptive residual connector, and a progressive dimensionalitygrouped bottleneck filter are incorporated to refine feature selection and reduce the model's complexity, boosting the model's overall performance. Experiments on vehicle lithium-ion battery datasets demonstrate that LDNet-RUL achieves state-of-the-art (SOTA) performance, with average RMSE of 0.23 and Score of 0.49. The model is lightweight, with only 7.92K parameters, achieves inference times of 3.94ms on GPU and CPU, and 5.39ms on the NVIDIA Jetson TX2.

This study aims to identify the key factors governing the thermal quenching of Mn⁴⁺ ion luminescence in fluoride-based phosphor materials used as red emitters in modern-day phosphor-converted LED devices. Here, we employ first-principles calculations for Mn⁴⁺-doped Na₂SiF₆, NaKSiF₆, and K₂SiF₆ hosts to explore how host properties and local coordination environments influence thermal quenching behavior. The ΔSCF method was used to model the geometric structures of the Mn⁴⁺⁴A₂ (ground) and ²E, ⁴T₂ (excited) states and the energies of the optical transitions between these states. Our results reveal that thermal quenching in Na₂SiF₆ and K₂SiF₆ phosphors occurs through thermally activated ²E → ⁴T₂ → ⁴A₂ crossover. In contrast, thermal quenching in NaKSiF₆ is due to other nonradiative decay pathways. Investigations of the mechanical stability of these fluorides show that NaKSiF₆ is mechanically unstable. We suggest that this property of the host limits the luminescence efficiency of the embedded Mn⁴⁺ ions. We also determined the reason for the difference in the intensity of the ²E → ⁴A₂ emission transition (ZPL) in the systems. These findings advance our fundamental understanding of the thermal quenching mechanism of Mn⁴⁺ ion luminescence in fluorides, and the results can aid future discoveries of technologically useful phosphors through high-throughput design methodologies.

Existing time series prediction models struggle with long-sequence data due to limitations in capturing long-range dependencies and processing multichannel information simultaneously. To address this gap, we first propose the channel independence bidirectional gated Mamba (CIBG-Mamba), a novel model enhancing gated recurrent neural networks by integrating Mamba into channel-independent recurrent units. At its core, CIBG-Mamba processes each data channel independently, with each channel utilizing three Mamba-enhanced recurrent units. This design improves memory capacity while enabling dynamic interchannel exchange, allowing the model to effectively handle multidimensional time series (MTS) data. Furthermore, an interactive recurrent mechanism (IRM) is introduced to optimize bidirectional information handling, employing channel-specific adaptive state transition functions. IRM significantly enhances the CIBG-Mamba's ability to capture complex time series dynamics across multiple dimensions. Finally, by fusing the above technologies, the prediction approach, i.e., the channel independence bidirectional gated Mamba with interactive recurrent mechanism (CIBG-Mamba-IRM) is constructed for time series forecasting. Additionally, a dataset with one year of data from five new energy vehicle air conditioners (NEVAC) is created. Through experiments on this dataset and five public time series datasets, our method proved effective in learning long-term dependencies and handling MTS data, achieving state-of-the-art results. Our code is available at https://github.com/XYouZheng/CIBG-Mamba.

Traditional treatment methods for tumors are inefficient and have severe side effects. At present, new therapeutic methods such as phototherapy, chemodynamic therapy, and gasodynamic therapy have been innovatively developed. High concentrations of hydrogen sulfide (H₂S) gas exhibit cancer-suppressive effects. Herein, a Prussian blue-loaded tetra-sulfide modified dendritic mesoporous organosilica (PB@DMOS) was rationally constructed with glutathione (GSH)-triggered/photothermal-enhanced H₂S signaling molecule release properties for gas therapy. The as-synthesized nanoplatform confined PB nanoparticles in the mesoporous structure of organosilica silica due to electrostatic adsorption. In the case of a GSH overexpressed tumor microenvironment, H₂S gas was controllably released. And the temperature increases due to the photothermal effects of PB nanoparticles, further enhancing H₂S release. At the same time, PB nanoparticles with excellent hydrogen peroxide catalytic performance also amplified the efficiency of tumor therapy. Thus, a collective nanoplatform with gas therapy/photothermal therapy/catalytic therapy functionalities shows potential promise in terms of efficient tumor therapy.

Isostatic pressure effects on the elastic and electronic properties of non-doped and Mn⁴⁺-doped K₂SiF₆ (KSF) have been investigated by first-principles calculations within density functional theory (DFT). Bulk modulus was obtained by the Murnaghan’s equation of states (EOS) using the relationship between volume and pressures at pressures between 0 and 40 GPa, and elastic constants were calculated by the stress–strain relationship giving small distortions at each pressure point. The other elastic parameters such as shear modulus, sound velocity and Debye temperature, which can be obtained from the elastic constants, were also estimated. The influence of external isostatic pressure on the electronic properties, such as crystal field strength 10Dq and emission energy of ²E → ⁴A² transition (E_em), of KSF:Mn⁴⁺ was also studied. The results suggest that 10Dq and E_em linearly increase and decrease, respectively, with increasing pressure.